论文部分内容阅读
提出一种直接优化求解控制量的轨迹跟踪控制策略，无须知道系统的先验知识，基于神经网络在线实时辨识，反复迭代求解已知跟踪轨迹的一步最优控制量。几组非线性系统的仿真表明该算法具有简捷快速，适应面宽的特点。
A direct optimal trajectory tracking control strategy for solving control variables is proposed. Without knowing the prior knowledge of the system, an iterative optimization algorithm based on neural network on-line real-time identification is used to solve the one-step optimal control of a known trajectory. Several groups of nonlinear system simulation shows that the algorithm has the characteristics of simple and fast, wide adaptation.